Management and treatment of contact lens-related Pseudomonas keratitis.

Pubmed and Medline were searched for articles referring to Pseudomonas keratitis between the years 2007 and 2012 to obtain an overview of the current state of this disease. Keyword searches used the terms "Pseudomonas" + "Keratitis" limit to "2007-2012", and ["Ulcerative" or "Microbial"] + "Keratitis" + "Contact lenses" limit to "2007-2012". These articles were then reviewed for information on the percentage of microbial keratitis cases associated with contact lens wear, the frequency of Pseudomonas sp. as a causative agent of microbial keratitis around the world, the most common therapies to treat Pseudomonas keratitis, and the sensitivity of isolates of Pseudomonas to commonly prescribed antibiotics. The percentage of microbial keratitis associated with contact lens wear ranged from 0% in a study from Nepal to 54.5% from Japan. These differences may be due in part to different frequencies of contact lens wear. The frequency of Pseudomonas sp. as a causative agent of keratitis ranged from 1% in Japan to over 50% in studies from India, Malaysia, and Thailand. The most commonly reported agents used to treat Pseudomonas keratitis were either aminoglycoside (usually gentamicin) fortified with a cephalosporin, or monotherapy with a fluoroquinolone (usually ciprofloxacin). In most geographical areas, most strains of Pseudomonas sp. (≥95%) were sensitive to ciprofloxacin, but reports from India, Nigeria, and Thailand reported sensitivity to this antibiotic and similar fluoroquinolones of between 76% and 90%.

Introduction

Microbial keratitis (MK), epithelial loss from the cornea with underlying stromal infiltration by white blood cells and disintegration of the stroma, occurs when one of the protective mechanisms of the ocular surface is disrupted. It is a vision-threatening condition that requires rapid and appropriate management and antibiotic treatment if vision loss is to be prevented. MK caused by Pseudomonas aeruginosa is commonly associated with contact lens wear (Table 1).1–21 Predisposing risk factors for microbial keratitis can vary with geographical location and can depend on the penetration of contact lens wear. The differences may also be associated with the incidence of single nucleotide polymorphisms (SNPs) in cytokine genes in different populations. Recently, SNPs in the gene for interleukin (IL)-10 have been associated with severity of and predisposition to MK.22 In developing countries, trauma to the eye may be a predominant risk factor,23 whereas in developed countries, contact lens wear is often the most important risk factor.24 A study from Malaysia suggested that as P. aeruginosa is also a common inhabitant of soil, water, and vegetation, it may also be the main pathogen following vegetation-related corneal injury in certain regions.15

Table 1

Percent of microbial keratitis cases associated with contact lens wear

Geographical location	Country	% MK associated with contact lens wear
North America	USA1	55
	USA2	26.5
South America	Brazil3	12.8
Europe	UK4	31
	UK5	32
	UK6	30.3
	Ireland7	41.1
	The Netherlands8	39.7
	Turkey9	3.2
	Italy10	46.1
Indian subcontinent	India11	17.14
	India12	8.2
Asia	Japan13	54.5
	Nepal14	0
	Malaysia15	21
	Thailand16	18.6
	Thailand17	32.4
Australasia	New Zealand19	29.4
	Australia20	21.7
	Australia21	21

The incidence of contact lens-related microbial keratitis has been estimated over the past 20 years, and has remained almost constant at 1/2500 contact lens wearers who wear lenses on a daily wear basis (that is removing lenses each night and placing in disinfecting solution prior to re-wearing the lens the next day), or 1/500 wearers if the lenses are worn on a continuous or extended wear basis (ie, the person wears lenses for 24 hours, sleeping in lenses overnight).25 It is now common for lens wearers to discard their lenses after 2 weeks or 1 month of wear.25 In a study from North America, it was found that the incidence of all ulcerative keratitis was 2.76 per 10,000 person-years (95% confidence interval [CI]: 2.46–3.09) but the incidence of contact lens-associated keratitis was 13.04 per 10,000 person-years (95% CI: 11.13–15.17), with an adjusted relative risk of 9.31 (7.42–11.7; P < 0.001) compared with non-contact lens wearers.1 Another study put the incidence of MK at 1.1 per 10,000 persons/year in the US24 but a different study found an incidence of 79.9 per 10,000 persons/year in Nepal.23 The risk with therapeutic contact lenses is higher at approximately 52/10,000 yearly.26 A study of armed forces of the UK evacuated because of keratitis from the Middle East showed an incidence of MK of 35 per 10,000 (with 74% being associated with soft contact lens wear).27

The percentage of microbial keratitis cases caused by Pseudomonas species (most likely P. aeruginosa) is shown in Table 2 for different geographical locations.1–13,15–18,28–44 Whilst P. aeruginosa/Pseudomonas sp. are usually a predominant causative agent, temperate zones tend to have a higher incidence of Gram-positive bacteria causing the disease and less aggressive keratitis.44 In most studies, Pseudomonas sp. are usually isolated in monoculture from cases of MK, however, a study from Thailand demonstrated that in 46% of MK cases caused by Pseudomonas sp. other Gram-negative bacteria including Escherichia coli, Acinetobacter calcoaceticus, Klebsiella pneumoniae, Serratia marcescens, and Enterobacter sp. could also be cultured.16 However, the predominance of P. aeruginosa during contact lens-associated MK is not always seen. For example, even though 29.4% of MK cases were associated with contact lens wear in a study from Wellington, New Zealand, no cultures of P. aeruginosa were reported.19 The predominant Gram-negative bacteria isolated was Moraxella sp. (12.5% of all bacterial isolates),19 and this predominance of Moraxella sp. from MK scrapes has been reported from a study in Christchurch, New Zealand.45 Climate may also affect the incidence of P. aeruginosa keratitis. In Australia, the incidence of P. aeruginosa contact lens microbial keratitis (CLMK) is increased in tropical compared to temperate zones, whereas the incidence of Serratia marcescens CLMK is higher in temperate zones.44

Table 2

Frequency of Pseudomonas sp. as a causative agent of microbial keratitis in different geographical regions

Geographical region	Country	Frequency (%) of Pseudomonas sp. as a total of all MK isolates
North America	USA1	0
	USA2	20.2
South America	Brazil3	12
	Brazil28	12 (41% of these caused by P. aeruginosa)
Europe	UK29	6 (1995–1998); 15 (2004–2007)
	UK4	12
	UK5	21
	UK6	28.5 (24.3% of total cases caused by P. aeruginosa)
	UK30	20.9
	Ireland7	33.3 (56.2 of CLMK)
	The Netherlands8	22.4
	Turkey9	6.6 (Pseudomonas sp.)
	Italy10	72.2
Middle East	Iraq31	42 (100% of those associated with contact lenses caused by Pseudomonas sp.)
	Kingdom of Bahrain32	54 (95% of those associated with contact lenses caused by P. aeruginosa)
	Various27 (predominantly Iraq)	71
	Oman33	28.8 (all CLMK)
Africa	Sierra Leone34	40
	Nigeria35	22.4
Indian subcontinent	India36	71 (only cases of CLMK examined, all Pseudomonas species were P. aeruginosa)
	India37	52
	India11	1
	India12	24.4
Asia	Japan38	2.8
	Japan18	20
	Japan13	1
	Thailand16	59
	Thailand17	55
	Malaysia15	58.6
	Hong Kong39	42.9 (85.7 of culture proven) for CLMK
	China40	20.07
	Taiwan41	47
Australasia	New Zealand42	3.4 (all P. aeruginosa)
	Australia43	8
	Australia21	17 (55% of these caused by P. aeruginosa)
	Australia44	35 (CLMK; 49.2 of culture proven cases)

Determinants of the clinical outcome of MK include distance of the ulcer from the limbus and the minimum inhibitory concentration (MIC) of the first antimicrobial used or lowest MIC if combination therapy was used.5 A large multicenter clinical trial with participants from India and the US has shown that P. aeruginosa ulcers were significantly worse for visual acuity than patients with other bacterial ulcers, but interestingly showed significantly more improvement in 3-month best-spectacle-corrected visual acuity than those with other bacterial ulcers.46 Pseudomonas sp. are often associated with the largest ulcers.5

Ideally, every case of presumed MK should be scraped for microbiological investigations, especially with the possibility of increasing isolation of antibiotic-resistant microbes. However, it must be borne in mind that there is often a small ulcer and so relatively little material might be obtained. Corneal scrapings obtained with a surgical blade (eg, Bard-Parker blade #15), Kimura spatula, or 21-gauge disposable needle should be inoculated on chocolate agar, sheep blood agar, and into thioglycolate broth, and incubated at 35°C. Sabouraud’s agar plates should also be used and these are maintained at 25°C to enhance fungal growth. Samples may also be inoculated onto non-nutrient agar and into brain heart infusion broth. Scraping of small lesions (smaller than 2.0 mm2) is probably not worthwhile, and patients with such lesions can be empirically treated. Scrapes should not only be sent for microbial culture, but also smeared onto microscope slides and examined by Gram stain (and potassium hydroxide if fungal keratitis is suspected). However, as there is often only a small amount of material, cultures on agar plates for bacteria and fungi, as well as Gram stain, are most often used. The following clinical parameters are useful in monitoring the clinical response to antibiotic therapy: blunting of the perimeter of the stromal infiltrate, decreased density of the stromal infiltrate, reduction of stromal edema and endothelial inflammatory plaque, reduction in anterior chamber inflammation, re-epithelialization, and cessation of corneal thinning.

Therapies used in different geographical locations are shown in Table 3.4,5,8,9,12,16,31,39,42,47–49 Monotherapy with ciprofloxacin (0.3%; or another fluoroquinolone) is commonly used. In severe cases, subconjunctival injections of gentamicin may be used.31 The combination of two fortified antibiotic preparations, 1.5% gentamicin and 5% cefuroxime, covers almost the entire range of common bacterial pathogens causing corneal ulcers. Randomized controlled trials have demonstrated that monotherapy with fluoroquinolones has non-inferiority and fewer side effects compared with combination therapy.50,51 A study from Iran recommended the concurrent use of ceftazidime and amikacin or ceftazidime and ciprofloxacin as the initial treatment based on antibiotic sensitivities of isolates, and as all P. aeruginosa isolates were resistant to chloramphenicol, trimethoprim, vancomycin, and cefazolin, these antibiotics should probably not be included in any empirical antibiotic regimen in that country.47 Data from Taiwan41 demonstrate that ciprofloxacin was statistically significantly more effective against P. aeruginosa than the combination of cefazolin and gentamicin. Whilst therapy is most often, if not always, commenced prior to results of cultures being obtained, a study from Japan has shown that the therapeutic outcome was better when antimicrobial agents were selected based on culture results, thus reemphasizing the importance of culture studies.18 Sometimes a combination of piperacillin/tazobactam might be effective with unresponsive P. aeruginosa MK.52

Table 3

Most common topical antimicrobial therapies used to treat Pseudomonas keratitis by geographical location

Geographical region	Country	Antibiotics commonly prescribed
Europe	The Netherlands8	Cefazolin and tobramycin/gentamicin; ofloxacin monotherapy
	Ireland9	Ceftazidime and vancomycin; ofloxacin
	UK4	Ciprofloxacin
	UK5	Ciprofloxacin or ofloxacin (84% monotherapy; 9% combination therapy)
Middle East	Iraq31	Ciprofloxacin
	Iran47	Fortified ceftazidime and vancomycin; ciprofloxacin for small (<2 mm) ulcers
Indian subcontinent	India12	Fortified cefazolin; tobramycin (modified depending on sensitivity analysis and clinical response)
Asia	Hong Kong39	Levofloxacin or gentamicin monotherapy; fortified gentamicin
	Thailand16	Fortified antibiotics (gentamicin or amikacin or ceftazidime and/or cefazolin); ciprofloxacin and/or tobramycin
Australasia	New Zealand48	Severe cases fortified gentamicin or Tobramycin; ciprofloxacin; mild cases ciprofloxacin; chloramphenicol
	New Zealand42	Fortified cefuroxime and tobramycin; ciprofloxacin in cases where scrape results show Gram-negative organisms resistant to tobramycin
	Australia49	Fluoroquinolone monotherapy; ceftazidime/gentamicin

The use of steroids in conjunction with antibiotics has been a source of controversy for many years, despite the demonstration in an animal trial that the combination of tobramycin and dexamethasone was safe and resulted in the reduction of clinical scores and lower bacterial numbers in the cornea.53 However, a recent large scale multicenter clinical trial that enrolled subjects in India and US found that the use of moxifloxacin combined with prednisolone phosphate did not improve overall clinical outcome.46

Sensitivity of Pseudomonas sp. to antibiotics by geographical region is shown in Table 4.2,7,8,16,20,21,28–31,35,37,40–42,47,54 Generally P. aeruginosa is sensitive to fluoroquinolones, but there have been reports of multi-resistant P. aeruginosa strains, for example, from Australia where the strains were resistant to ciprofloxacin, gentamicin, tobramycin, and amikacin but was sensitive to ceftazidime, imipenem, meropenem, and timentin.55 Recent data examining possible synergistic activity between different classes of antibiotics against P. aeruginosa has shown that a combination of meropenem/ciprofloxacin gave the lowest mean fractional inhibitory concentrations (ie, best synergy) for P. aeruginosa isolates, with 90% of isolates showing an additive or synergistic effect56 and so this may be a promising therapy for the more resistant strains.

Table 4

Sensitivity to antibiotics of Pseudomonas sp. in different geographical regions

Pseudomonas type	Country	Percentage of strains sensitive to antibiotic
		Ciprofloxacin	Gentamicin	Cephalosporin	Tobramycin	Chloramphenicol
P. aeruginosa	USA2	100 (levofloxacin = 100)	93.7a	ND	93.7a	ND
P. aeruginosa	Brazil28	100 (ofloxacin = 100; gatifloxacin = 100)	97	ND	100	ND
Pseudomonas sp.	Brazil54	95 (ofloxacin = 95; gatifloxacin = 95)	ND	ND	ND	ND
Pseudomonas sp.	Ireland7	100 (ofloxacin = 100)	100	73 (cefotaxime); 100 (ceftazidime); 18 (cefuroxime)	ND	ND
P. aeruginosa	UK30	98.6 (levofloxacin = 99.3; moxifloxacin = 100)	96.4	99.3 (ceftazidime)	ND	ND
Pseudomonas sp.	UK29	100	100	100 (1995–1998); 0 (2004–2007) (cefuroxime)	ND	ND
P. aeruginosa	The Netherlands8	100	ND	ND	ND	ND
Pseudomonas sp.	Iraq31	62	55	2 (cefazolin)	ND	0
P. aeruginosa	Iran47	100	93	0 (cefalozin); 100 (ceftazidime)	ND	3
P. aeruginosa	India37	85 (norfloxacin = 82; ofloxacin = 87; gatifloxacin = 88; moxifloxacin = 79)	33	0 (cefalozin); 64 (cephotaxime); 80 (cetazidime)	30	60
P. aeruginosa	Nigeria35,b	90 (ofloxacin = 80)	90	20 (cephalexin)	ND	10
P. aeruginosa	Taiwan41	99	91	99 (ceftazidime)	ND	ND
P. aeruginosa	Thailand16	100 (data for ofloxacin)	100	100 (ceftazidime)	ND	ND
Pseudomonas sp.	China40	76 (ofloxacin = 89; levofloxacin = 96)	ND	ND	87	ND
P. aeruginosa	New Zealand42	99c	ND	99.7 (cefuroxime)	100	ND
P. aeruginosa	Australia20	100	100	ND	ND	100
P. aeruginosa	Australia21	100	100	100 (ceftazidime or cefotaxime)	ND	ND

Notes:

Data supplied as ‘intermediate or resistant to gentamicin or tobramycin’;

all ocular infections not just MK;

data supplied for all Gram-negative microbes combined.

Abbreviations: MK, microbial keratitis; ND, not given or determined.

Comparisons between Tables 3 and 4 demonstrate that ciprofloxacin is the most commonly prescribed antibiotic to treat MK in Iraq, however only 62% of Pseudomonas sp. are sensitive to it. Likewise for India, tobramycin is one of the most commonly prescribed antibiotics but only 30% of Pseudomonas sp. are sensitive to it. This is different from all other most commonly prescribed treatments in other geographical locations which are >95% effective. Whilst there are no true cut-off points for sensitivity or resistance for topically applied antibiotics, it is perhaps important for those countries where there are high levels of apparently resistant strains of P. aeruginosa to monitor the clinical outcome of MK very carefully.

In conclusion, Pseudomonas sp. (predominantly P. aeruginosa) is often isolated from cases of contact lens-induced microbial keratitis. The most commonly used therapies to treat this disease are either monotherapy with a fluoroquinolone or fortified aminoglycosides. Strains of P. aeruginosa isolated from contact lens-induced MK are commonly still sensitive to these antibiotics, but geographic differences in sensitivity exist and should be taken into account when recommending treatment options.